PROJECT ABSTRACT The discovery of genetic risk factors for alcohol and tobacco use and dependence has largely focused on the contribution of individual genetic markers, such as single nucleotide polymorphisms (SNPs). While this approach has been useful in mapping the genetic liability of rare diseases with a Mendelian pattern of inheritance, the approach has had limited success with complex diseases characterized by a more polygenetic architecture. The identification of a set of genetic factors that account for individual differences in the liability to use and misuse alcohol and tobacco/nicotine will be a major step in understanding mechanisms of undercontrolled use. Moreover, the development of novel resources to predict or infer risk for problematic use of alcohol and tobacco based on genetic factors that largely explain individual differences in alcohol and tobacco use/problems, as well as other behavioral or neurocognitive phenotypes, will be a major step in translating research discoveries into prevention strategies. The current proposal addresses the problem of limited predictive power in alcohol and tobacco/nicotine genetic studies by proposing novel research that uses an integrative approach and existing data and bioinformatics resources to: (1) localize genetic variants that comprise the additive genetic effects on alcohol and tobacco use and dependence, and (2) developing a novel and freely-available resource that enhances the way existing genetically informed samples are used for alcohol and tobacco risk prediction. The current application assembles a multidisciplinary team of molecular genetics and computational and statistical geneticists to execute these project aims.